Youtu Parsing Omni
Parameters
8B
Context Length
1.05M
Modality
Text
Architecture
Dense
License
other
Release Date
9 Oct 2026
Knowledge Cutoff
-
API Pricing (per 1M)
Self-hosted only
System Requirements
VRAM requirements for different quantization methods and context sizes
1,024 tokens
Consumer
1x RTX 4090
24GB VRAM
Datacenter
1x NVIDIA A100
80GB VRAM
Apple Silicon
1x Apple M3 Max
128GB VRAM
1,048,576 tokens
Consumer
5x RTX 4090
24GB VRAM
Datacenter
2x NVIDIA A100
80GB VRAM
Apple Silicon
1x Apple M3 Max
128GB VRAM
Architecture Diagram
Evaluation Benchmarks
No evaluation benchmarks for Youtu Parsing Omni available.
Rankings
Overall Rank
-
Coding Rank
-
About Youtu Parsing Omni
Youtu Parsing Omni is Tencent Youtu Lab's multimodal parsing model built on the VITA vision-language architecture. It specializes in fine-grained document understanding, visual layout analysis, and omni-modal information extraction.
Technical Specifications
Attention
Attention Structure
Multi-Layer Attention
Attention Heads
32
Key-Value Heads
8
Attention Head Dimension
64
Position Embedding
ROPE
RoPE Theta
1,600,000
Sliding Window Attention
No
Sliding Window Size
-
Sliding Window Ratio
-
Linear Attention
No
Linear Attention Ratio
-
KV Latent Dimension
576
Normalization
RMS Normalization
Activation Function
SwigLU
Dimensions
Auxiliary Parameters
-
Hidden Dimension Size
2,560
Number of Layers
40
FFN Intermediate Size (Dense)
9,728
Multi-Token Prediction Heads
0
Tokenizer
Vocabulary Size
133,632
Resources
About Hunyuan
Tencent Hunyuan large language models with various capabilities.